Stanford Research, packed with 797's. Bear are you still in Hannacroix I have a few tubes of parts for you?
Okay, and good. But, how exactly? Cymbals? Ample training and practice? Finding the best of best test subjects? Point me to the papers, please if you would.
Please, to clarify, the HEDD A/D not intended or used as an effect of any kind. Only the optional digital processing is intended to be an effect, which I don't use.
The discussion of it here is applicable to the issue of recording accurate wave files of opamp test boxes. That's all.
This would be counterproductive in generating any useful information for the population at large. That would be like using the abilities of an idiot savant to make decisions about how to best spend money on education. I suggested this years ago, the possibility of auditory savants that is. Funny thing people with perfect pitch usually end up considering it a curse.
I think it's helpful to know both the limits and the averages. Certainly we have information like that for IQ because we test large numbers of people. We now have people doing superforecasting because a large enough program found exceptional people who could do something never thought possible before. And I suspect that some people are being treated very badly because we seem to be so reluctant to look at limits.
Good start on that post but it derails with the 'my grado headphones are better than my speakers, everyone should use headphones'
Grado are not source or amp quality revealing speakers, they are very good dynamic headphones.
Sennheisers have way more sensitive to the source material and output chain.
, I prefer my sound system way more than the headphones, more stereo, more depths of sound, more airiness and quite sensitive considering the easiness of listening.
Anyone should do the test with their most distortion revealing sound system, headphone, car, plane, boat, or just good old stereo.
the art of a recording engineer/producer:
1. compress the sound as much as possible
2. make sure it sounds great on walkmans / beach radios
3. make double sure any sound system with 20 db dynamic range will handle the crappy mess
Maybe a little exaggerating, I am not a direct to vinyl fan, however the less the engineers mess with the sound usually the better on hifi systems.
1. compress the sound as much as possible
2. make sure it sounds great on walkmans / beach radios
3. make double sure any sound system with 20 db dynamic range will handle the crappy mess
Maybe a little exaggerating, I am not a direct to vinyl fan, however the less the engineers mess with the sound usually the better on hifi systems.
I know it gets very freak show at some level but it would be interesting for these savants to show off their abilities like magicians or the guys that can play 50 chess games at once blindfolded. I have a sneaking suspicion that virtually no one will come forward.
How about some Joy Division, The Jam, PIL, Anti-Nowhere League, Nurse With Wound, Mark Pauline, Angry Samoans, Discharge, Iggy Pop, Cockney Rejects, Jimmy Pursey, Diamanda Galas, anything but the Beatles.
That dual differential discrete op-amp looks very good! I wish that I could try one of those. This is similar to what I do with discrete designs, but at much greater cost.
I replied about some distortion of an amp in direct response to your comments about double digit distortion. This in the context of talking about hearing distortion and in particular distortion that is objectionable. I'm not sure where you were going on that, but I seem to recall you were talking about distortion that people like.
If it would help to refocus what we are talking about with less of the various side excursions, I was previously talking about learning to hear distortion because it was objectionable to me. And that included not liking something about the Lynx2 A/D because of apparent distortion. Who cares? If we are going to find out what's wrong with opamps as various people have been discussing using some kind of listening test, there was the send hardware option, and the use software and wave file version. For the latter, it would appear best to use unobjectionably good data converters to make test files, and good DACs for test listening so that subtle opamp distortion has a chance of being heard.
Last edited:
Scott- Thanks for digging out that early Dafos stuff. That was a long time ago.
Bear- They lifted the whole array (i think called the beast?) and dropped it on the stage. Its the stage moving that generates the low frequency component.
The recorder you may have seen was the early 3M M23 I built for Keith with a 3 track head and the custom electronics we made. Its 1/4" also. They are so big people tend to scale them even larger in memory. Heavy also.
Are we going to now open a side topic on how chess masters do what you just said and how it is different from listening? We could do that, but please bear in mind how easily these things can take on a life of their own.
Anyway, maybe no one would come forward for testing because we don't know how to test for limits very well. We evidently have experience at trying to measure average. It's not clear that the same testing for both is optimal. It certainly isn't for IQ. High IQ sub-populations are best measured with specialized testing for that purpose.
Last edited:
Opamp performance can vary according to nonlinearities and loading modes of the input signal and further the current capability of the DUT opamp output stage to drive the combinations of output and feedback network loadings.
Scott's circuit illustrated above elegantly sidesteps these issues/dependencies.
This configuration gives the ability to position swap similar opamps and presumably highlight any subjective differences.....or roll DUT's into either circuit position or both.
So building stereo stages into suitable sized diecast enclosures (peek and tamper proofable) with mirrored circuit positions ought to be a tighter test of actual opamp differences and substantially reduce loading effects.
My two bob.
Dan.
Scott's circuit illustrated above elegantly sidesteps these issues/dependencies.
This configuration gives the ability to position swap similar opamps and presumably highlight any subjective differences.....or roll DUT's into either circuit position or both.
So building stereo stages into suitable sized diecast enclosures (peek and tamper proofable) with mirrored circuit positions ought to be a tighter test of actual opamp differences and substantially reduce loading effects.
My two bob.
Dan.
I have a Lynx L22 which is essentially the same card with 2 in and 2 out. It uses the AK5394A ADC (and so does Protools). In my experience the opamps and other external parts are more likely to limit the distortion. I can make some measurements to see what its doing. Their input circuit is different from that on the EMU cards using the same ADC.
I have read that the recording crowd preferred the RME stuff to the Lynx stuff.
I really don't know what an HEDD is. I could not find it online.
That's a composite amplifier circuit and isn't going to be stable just rolling random op-amps in. I mean... come on. Probably does sound different if you build an oscillator.
HEDD webpage: hedd
Lynx now makes, I think, the Aurora line which is based on Lynx2 architecture. It is apparently very popular. The Lynx converters do sound better than most at their price point. It's just that the A/D in the HEDD is better than my Lynx2, and the Benchmark DAC-1 is better than the Lynx2 D/A. Why mention this? Because what it means to me is that if we are trying to see if we can hear opamp distortion using wave files, then even something the quality of the Lynx2 is probably not as well suited for wave recording/playback as the HEDD/DAC-1 combination, because the masking distortion in the latter is, I believe, less than for the Lynx, and therefore better suited to testing opamps for audible distortion. In other words, I am tying to emphasize that using listening tests to see what's wrong with opamps works best if you have top quality coverters, are well trained to hear distortion, and you keep in practice. That's it.
Last edited:
Add to that a bunch of similar Aussie stuff and we have comparable misspent youth. 😱 .
Yup, for crissakes no Beetles, no Dire Straights, no Steely Dan, no Pink Floyd even.
The files that Mooly has put up in the past just bored me or drove me out of the room so much so that I didn't bother to complete the testings.
Suggestions and concensus on a wide set of choices of suitable genres and tracks would be one step in this study.
Dan.
As stated in this thread previously, the A/B audio quality test concept is very difficult. For example see pg 31 of this paper by Simon Liljestrom at Uppsala.
http://uu.diva-portal.org/smash/get/diva2:410669/FULLTEXT01.pdf
-
http://uu.diva-portal.org/smash/get/diva2:410669/FULLTEXT01.pdf
-
- Status
- Not open for further replies.